Automatic Machine Learning to Differentiate Pediatric Posterior Fossa Tumors on Routine MR Imaging.
Journal
AJNR. American journal of neuroradiology
ISSN: 1936-959X
Titre abrégé: AJNR Am J Neuroradiol
Pays: United States
ID NLM: 8003708
Informations de publication
Date de publication:
07 2020
07 2020
Historique:
received:
30
11
2019
accepted:
30
04
2020
entrez:
15
7
2020
pubmed:
15
7
2020
medline:
1
12
2020
Statut:
ppublish
Résumé
Differentiating the types of pediatric posterior fossa tumors on routine imaging may help in preoperative evaluation and guide surgical resection planning. However, qualitative radiologic MR imaging review has limited performance. This study aimed to compare different machine learning approaches to classify pediatric posterior fossa tumors on routine MR imaging. This retrospective study included preoperative MR imaging of 288 patients with pediatric posterior fossa tumors, including medulloblastoma ( For 3-way classification, the radiomics model by automatic machine learning with the Tree-Based Pipeline Optimization Tool achieved a test micro-averaged area under the curve of 0.91 with an accuracy of 0.83, while the most optimized model based on the feature-selection method χ Automatic machine learning based on routine MR imaging classified pediatric posterior fossa tumors with high accuracy compared with manual expert pipeline optimization and qualitative expert MR imaging review.
Sections du résumé
BACKGROUND AND PURPOSE
Differentiating the types of pediatric posterior fossa tumors on routine imaging may help in preoperative evaluation and guide surgical resection planning. However, qualitative radiologic MR imaging review has limited performance. This study aimed to compare different machine learning approaches to classify pediatric posterior fossa tumors on routine MR imaging.
MATERIALS AND METHODS
This retrospective study included preoperative MR imaging of 288 patients with pediatric posterior fossa tumors, including medulloblastoma (
RESULTS
For 3-way classification, the radiomics model by automatic machine learning with the Tree-Based Pipeline Optimization Tool achieved a test micro-averaged area under the curve of 0.91 with an accuracy of 0.83, while the most optimized model based on the feature-selection method χ
CONCLUSIONS
Automatic machine learning based on routine MR imaging classified pediatric posterior fossa tumors with high accuracy compared with manual expert pipeline optimization and qualitative expert MR imaging review.
Identifiants
pubmed: 32661052
pii: 41/7/1279
doi: 10.3174/ajnr.A6621
pmc: PMC7357647
doi:
Types de publication
Journal Article
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
1279-1285Subventions
Organisme : NCI NIH HHS
ID : F30 CA239407
Pays : United States
Organisme : NIBIB NIH HHS
ID : T32 EB001680
Pays : United States
Informations de copyright
© 2020 by American Journal of Neuroradiology.
Références
Front Oncol. 2019 Aug 22;9:806
pubmed: 31508366
Radiology. 2016 Feb;278(2):563-77
pubmed: 26579733
AJNR Am J Neuroradiol. 2010 Oct;31(9):1613-6
pubmed: 20538820
Neuroimaging Clin N Am. 1999 Nov;9(4):671-90
pubmed: 10517939
Cancer Cell. 2017 Mar 13;31(3):326-341
pubmed: 28292436
Neurooncol Pract. 2019 Dec;6(6):428-437
pubmed: 31832213
Pac Symp Biocomput. 2018;23:192-203
pubmed: 29218881
Neuroradiol J. 2014 Apr;27(2):233-44
pubmed: 24750714
AJNR Am J Neuroradiol. 2006 Jun-Jul;27(6):1362-9
pubmed: 16775298
AJNR Am J Neuroradiol. 2014 May;35(5):1009-15
pubmed: 24309122
Neuroimaging Clin N Am. 2017 Feb;27(1):1-37
pubmed: 27889018
J Magn Reson Imaging. 2012 Jan;35(1):32-47
pubmed: 21989968
Sci Rep. 2015 Aug 17;5:13087
pubmed: 26278466
Neuroradiology. 2014 Sep;56(9):781-8
pubmed: 24974083
Jpn J Radiol. 2017 Aug;35(8):448-453
pubmed: 28550357
BMC Med Res Methodol. 2019 Mar 19;19(1):64
pubmed: 30890124
Neurooncol Pract. 2018 Mar;5(1):18-27
pubmed: 29692921
J Neuroradiol. 2020 Feb;47(1):46-53
pubmed: 31541639
Pediatr Neurosurg. 2004 Jan-Feb;40(1):8-15
pubmed: 15007223
Ir Med J. 2001 Feb;94(2):52-3
pubmed: 11321174
Radiology. 2020 May;295(2):328-338
pubmed: 32154773
J Pediatr Neurosci. 2017 Jul-Sep;12(3):245-250
pubmed: 29204199
Front Oncol. 2019 Nov 05;9:1164
pubmed: 31750250
J Neurooncol. 2012 May;108(1):163-71
pubmed: 22350379
Sci Rep. 2017 Aug 31;7(1):10117
pubmed: 28860628
Acad Radiol. 2012 Jul;19(7):794-800
pubmed: 22513110
Bioinformatics. 2020 Jan 1;36(1):250-256
pubmed: 31165141
Pac Symp Biocomput. 2018;23:460-471
pubmed: 29218905
Eur Radiol Exp. 2018 Nov 14;2(1):36
pubmed: 30426318
J Magn Reson Imaging. 2007 Dec;26(6):1390-8
pubmed: 17968955
Neuro Oncol. 2020 Mar 5;22(3):393-401
pubmed: 31563963